• State TI Thinks Resolved
  • Locked Locked
  • Replies 13 replies
  • Answers 4 answers
  • Subscribers 46 subscribers
  • Views 5658 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

ADS1298ECGFE-PDK: ADS1298ECGFE-PDK

Ajinkya Gorhe
Prodigy 80 points
Part Number: ADS1298ECGFE-PDK
Other Parts Discussed in Thread: ADS1298, ADS1298R

Hello,

I am using ADS1298 for 12 lead ECG measurement.

The front-end circuit design of the ECG on our custom board is done with reference to the development board circuit of ADS1298R.

 

We have tested the custom board with the ECG simulator and the waveforms are satisfactory , similar to the waveforms captured on development board. But when the same board is connected to human body we are not able to get any waveforms.

We are getting only noise.

 

We have tested the ADS1298R reference board on human body and the observation is similar. We are getting noise there as well.

 

Can somebody please help with this problem ?

Please let me know if you need more details on this.

  • 0
    TI__Mastermind 38745 points
    Hello Ajinkya,

    The main difference between a simulator and the actual patient is the impedance of the electrode-skin barrier. In many cases, the impedance can be in the tens or hundreds of kOhm and in the MOhms for dry electrodes. This high impedance makes it easy for utility interference to couple on to the electrodes from overhead lights, fans, motors, etc. Can you post a photo of the data you are getting? Does it have a lot of mains interference? If that is the case, there are a couple of things you can do, but let's make sure that is what is happening before I advise you on that.

    Brian
  • 0 in reply to Brian Pisani
    Prodigy 80 points

    Hello Brian,

    Thanks for the quick help.

    Please find the attached wave-forms captured with Simulator and actual patient.

    For the simulator all 10 probes were connected and with body only RA and LA were connected. So please ignore other wave-forms. 

    Please let me know if you need more details.

    ECG_body.xlsxECG_Simulator.xlsx

  • 0 in reply to Ajinkya Gorhe
    Prodigy 80 points

    Hello Brian,

    I have captured the wave-forms with the reference board in a dark room. Still the problem is retained.

    Please find the attached images of the wave-forms with the simulator and human body captured on the reference board.

  • 0 in reply to Ajinkya Gorhe
    TI__Mastermind 38745 points

    Hello Ajinkya,

    It seems like in both cases the noise is from mains interference. This can occur even if lights are off in a particular room because the signal can come from other places. This problem is faced by every ECG acquisition device. There are two ways that other customer handle this problem:

    1. Using the Right Leg Drive amplifier on the ADS1298 to form a closed loop common-mode cancellation system. You can find out more about this concept in this app note.
    2. Digital notch filter. In many cases even with using the RLD, customers implement a 50 or 60 Hz notch filter in their microprocessor or DSP.

    Using wet electrodes may also improve things, but that may not be an option for your product.

    Brian

  • 0 in reply to Brian Pisani
    Prodigy 80 points

    Hello Brian,

    Actually the wave-forms are captured using Wet electrodes of 3M make. And to avoid mains interference I had disconnected the laptop charger and simulator was also working on battery. So what could be the possible reason of this noise ?

    As you have suggested I will try to use RLD amplifier and will let you know the results.

    Thanks,

    Ajinkya

  • 0 in reply to Ajinkya Gorhe
    TI__Mastermind 38745 points
    Hello Ajinkya,

    The interference couples on to the patient from all over the place. I'm not sure you could ever avoid it in the actual application environment. Right leg excitation is the most common way to reduce this type of interference. Please let me know the results of your experiments.

    Brian
  • 0 in reply to Brian Pisani
    Prodigy 80 points

    Hello Brian,

    I have tried using RLD amplifier with the reference board using utility provided with the development board. 

    But I am not getting any waveform with the settings I am doing. Please check the settings with the attached snapshot and let me know where I am going wrong.

  • 0 in reply to Ajinkya Gorhe
    TI__Mastermind 38745 points
    Ajinkya,

    Set the channel inputs to "Normal" and then in the "LOFF and RLD" tab, select which inputs to use for right leg drive inputs. On the EVM you can at least use all the positive inputs and IN2N and IN3N.

    Brian
  • 0 in reply to Brian Pisani
    Prodigy 80 points

    Hello Brian,

    I have tried with RLD amplifier and I am getting much better signal which is closer to expected signal. But this signal is not as clean as the one obtained from simulator. Observing the FFT , it is found that  the signal still has some 50Hz components and its harmonics. Seems that I have to use a Notch filter to get rid of this 50Hz component.

    Can you help with any sample Matlab code of such filter with which I can check this ? 

    Thanks,

    Ajinkya 

  • 0 in reply to Ajinkya Gorhe
    TI__Mastermind 38745 points
    Hello Ajinkya,

    I am glad the performance has improved! Unfortunately I am not the expert on developing DSP code. The task of determining the type and performance of a digital notch filter is outside of the scope of the service we provide on this forum. Perhaps one of your colleagues has developed something similar in the past which you could leverage.

    Brian
  • 0 in reply to Brian Pisani
    Prodigy 80 points

    Hello Brian,

    Please find the attached sheet for the ECG wave-forms captured on body after using RLD.

    Seems that the mains 50Hz noise is eliminated to a greater extend after using RLD and the irregularities in the wave-from are because of the problem in captured data.

    The problem seems that the amplitude of the T wave is on a higher side and the P wave looks attenuated. 

    Can you please examine these wave-forms and let me know the cause of the problem ? I have tried applying a 50Hz notch filter and did not observe much difference in performance. 

    Thanks,

    Ajinkya Board-1-ECG.xlsx

  • 0 in reply to Brian Pisani
    Prodigy 80 points

    Hello Brian,

    Please find the attached wave-forms for the ECG signal captured on human body. 

    The wave-forms are much better now ; but some noise is still present. It is also observed that amplitude of T wave is large and amplitude of P wave is very small.

    I have tried applying 50 Hz notch filter on these signals but did not see any improvement with that. It seems that the problems are with the captured data itself.

    Can you please check the wave-forms and let me know where the problem is ?2604.Board-1-ECG.xlsx

  • 0 in reply to Ajinkya Gorhe
    TI__Mastermind 38745 points
    Hello Ajinkya,

    Yes the 50 Hz interference has been reduced significantly and the ECG waveform is coming in clearly. I can't be sure that the size of the T wave is a "problem". The size of the various parts of the ECG waveform will depend on which ECG lead is used, the placement of the electrodes, and the patient. My theory is that this is an accurate recording of the patient's ECG. The ADC itself does not perform any ECG specific filtering that would increase the magnitude of one part of the QRS complex versus another. Do the characteristics change when the electrodes are moved, a different lead is used, or a different patient is tested?

    One way to test this theory would be to record the patient's ECG waveform with an existing ECG recorder using the same electrode placement and see what the signal looks like.

    Brian